| 基于PrefixSpan的快速交互序列模式挖掘算法 |
| |
| 引用本文： | 陆介平,刘月波,倪巍伟,刘同明,孙志挥.基于PrefixSpan的快速交互序列模式挖掘算法[J].东南大学学报(自然科学版),2005,35(5):692-696. |
| |
| 作者姓名： | 陆介平 刘月波 倪巍伟 刘同明 孙志挥 |
| |
| 作者单位： | 东南大学计算机科学与工程系,南京,210096;江苏科技大学电子与信息学院,镇江,212003;上海工程技术大学科研处,上海,200336;东南大学计算机科学与工程系,南京,210096;江苏科技大学电子与信息学院,镇江,212003 |
| |
| 基金项目： | 国家自然科学基金，江苏省自然科学基金 |
| |
| 摘 要： | 为了克服序列模式挖掘过程中重复运行挖掘算法而产生的时空消耗,提出了一个快速、简单而有效序列模式的交互式算法FISPM,利用前次挖掘得到的序列构造序列模式数据库用来存储挖掘出来的所有序列, 通过缩减本次挖掘所要构造投影数据库的频繁项的数量来减少构造投影数据库所需的时间以及投影数据库的大小,从而减少时间和空间消耗,提高挖掘效率.通过设置全局最小支持度来减少算法迭代次数. 实验结果证明在交互挖掘过程中FISPM效率优于PrefixSpan.
|
| 关 键 词： | 数据挖掘 序列模式 交互式挖掘 投影数据库 |
| 文章编号： | 1001-0505（2005）05-0692-05 |
| 收稿时间： | 04 15 2005 12:00AM |
| 修稿时间： | 2005-04-15 |
Fast interactive sequential pattern mining algorithm based on PrefixSpan |
| |
| Lu Jieping,Liu Yuebo,Ni Weiwei,Liu Tongming,Sun Zhihui.Fast interactive sequential pattern mining algorithm based on PrefixSpan[J].Journal of Southeast University(Natural Science Edition),2005,35(5):692-696. |
| |
| Authors: | Lu Jieping Liu Yuebo Ni Weiwei Liu Tongming Sun Zhihui |
| |
| Institution: | 1 Department of Computer Science and Engineering, Southeast University, Nanjing 210096, China;2 School of Electronics and Information, Jiangsu University of Science and Technology, Zhenjiang 212003, China;3 Scientific Research Office, Shanghai University of Engineering Science,Shanghai 200336, China |
| |
| Abstract: | A novel sequential patterns mining method based on PrefixSpan,called FISPM(fast interactive sequential patterns mining algorithm),is proposed in this paper to reduce time-consuming in rerunning algorithm of sequential patterns query.In which,by building sequence patterns base(SPB) for saving all the mined patterns and minimum support,the number of frequent items of the projection databases constructed by the correct mining which based on the previously mined sequences can be reduced,and the interactive process and the efficiency can be spended up by using SPB to reduce the time and space consuming of projected database.Furthermore,using globalthreshold can reduce the number of iteration.The results of experiments on several different databases indicate that the performance of FISPM is better than that of PrefixSpan in the interactive mining. |
| |
| Keywords: | data mining sequential patterns interactive mining projection database |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
